Terminal device, method and computer readable medium for communication
The location server facilitates PRU availability determination for UE-based positioning systems, addressing accuracy issues in AI/ML model performance monitoring by providing PRU information, thus enhancing the reliability of UE-based positioning.
Patent Information
- Application Number
- PCT/CN2024/108184
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-07-29
- Publication Date
- 2026-02-05
AI Technical Summary
Existing UE-based positioning systems using AI or ML models face challenges in accurately monitoring model performance due to the unavailability of positioning reference units (PRUs) with known locations, especially in non-line of sight (NLOS) conditions, which affects the accuracy of ground truth labeling.
A location server determines the availability of PRUs for model performance monitoring by receiving reference outputs from terminal devices and transmitting relevant PRU information or locations, enabling effective model performance monitoring even in general situations.
This approach allows for accurate model performance monitoring in various environments, ensuring reliable operation of AI or ML models in UE-based positioning systems.
Smart Images

Figure CN2024108184_05022026_PF_FP_ABST
Abstract
Description
TERMINAL DEVICE, METHOD AND COMPUTER READABLE MEDIUM FOR COMMUNICATIONTECHNICAL FIELD
[0001] Embodiments of the present disclosure generally relate to the field of telecommunication, and in particular, to a terminal device, method and computer readable medium for communication.BACKGROUND
[0002] User equipment (UE) -based positioning may be performed with an artificial intelligence (AI) or machine learning (ML) model at UE side to obtain a predicted location of a UE. In label based model performance monitoring methods for UE-based positioning with the AI or ML model at UE side, a ground truth label (i.e., target UE location) is provided for monitoring accuracy of outputs of the AI or ML model. To perform model performance monitoring in a general situation, a comprehensive procedures should be designed.SUMMARY
[0003] In general, example embodiments of the present disclosure provide a location server, a terminal device, methods and computer readable medium for communication.
[0004] In a first aspect, there is provided a location server. The location server comprises a processor. The processor is configured to cause the location server to: receive, from a terminal device, at least one first reference output of at least one artificial intelligence (AI) or machine learning (ML) model; determine a second reference output of the at least one AI or ML model based on the at least one first reference output; determine, based on the second reference output, whether a positioning reference unit (PRU) associated with the terminal device is available for model performance monitoring of the at least one AI or ML model at the terminal device; and transmit, to the terminal device, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.
[0005] In a second aspect, there is provided a terminal device. The terminal device comprises a processor. The processor is configured to cause the terminal device to: receive, from a location server, at least one reference location where a reference positioning reference unit (PRU) is available or information about the reference PRU is available or model performance monitoring of at least one artificial intelligence (AI) or machine learning (ML) model is applicable; based on determining that one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, transmit, to the location server, a request for the model performance monitoring; transmit the at least one first reference output to the location server; and based on determining that information about a PRU associated with the terminal device is received from the location server or the PRU, perform the model performance monitoring based on the information about the PRU.
[0006] In a third aspect, there is provided a method for communication. The method comprises: receiving, from a terminal device, at least one first reference output of at least one artificial intelligence (AI) or machine learning (ML) model; determining a second reference output of the at least one AI or ML model based on the at least one first reference output; determining, based on the second reference output, whether a positioning reference unit (PRU) associated with the terminal device is available for model performance monitoring of the at least one AI or ML model at the terminal device; and transmitting, to the terminal device, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.
[0007] In a fourth aspect, there is provided a method for communication. The method comprises: receiving, from a location server, at least one reference location where a reference positioning reference unit (PRU) is available or information about the reference PRU is available or model performance monitoring of at least one artificial intelligence (AI) or machine learning (ML) model is applicable; based on determining that one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, transmitting, to the location server, a request for the model performance monitoring; transmitting the at least one first reference output to the location server; and based on determining that information about a PRU associated with the terminal device is received from the location server or the PRU, performing the model performance monitoring based on the information about the PRU.
[0008] In a fifth aspect, there is provided a computer readable medium having instructions stored thereon. The instructions, when executed on at least one processor of a device, cause the device to perform the method according to the third aspect or the fourth aspect.
[0009] It is to be understood that the summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Through the more detailed description of some embodiments of the present disclosure in the accompanying drawings, the above and other objects, features and advantages of the present disclosure will become more apparent, wherein:
[0011] Fig. 1A illustrates an example communication network in which embodiments of the present disclosure can be implemented;
[0012] Fig. 1B illustrates another example communication network in which embodiments of the present disclosure can be implemented;
[0013] Fig. 1C illustrates a signaling chart illustrating an example process for Location Service Support by NG-RAN in accordance with some embodiments of the present disclosure;
[0014] Fig. 1D illustrates a signaling chart illustrating an example process for SLPP PDU transfer between UEs in accordance with some embodiments of the present disclosure;
[0015] Fig. 2 illustrates a signaling chart illustrating an example process for communications in accordance with some embodiments of the present disclosure;
[0016] Fig. 3 illustrates an example of determining a PRU associated with the terminal device in accordance with some embodiments of the present disclosure;
[0017] Figs. 4, 5 and 6 illustrate a signaling chart illustrating an example process for communications in accordance with some embodiments of the present disclosure, respectively;
[0018] Fig. 7 illustrates an example of reference locations in accordance with some embodiments of the present disclosure;
[0019] Fig. 8 illustrates a signaling chart illustrating an example process for communications in accordance with some embodiments of the present disclosure;
[0020] Fig. 9 illustrates a flowchart of an example method in accordance with some embodiments of the present disclosure;
[0021] Fig. 10 illustrates a flowchart of an example method in accordance with some embodiments of the present disclosure; and
[0022] Fig. 11 is a simplified block diagram of a device that is suitable for implementing embodiments of the present disclosure.
[0023] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION
[0024] Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitations as to the scope of the disclosure. The disclosure described herein can be implemented in various manners other than the ones described below.
[0025] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
[0026] As used herein, the term “terminal device” refers to any device having wireless or wired communication capabilities. Examples of the terminal device include, but not limited to, user equipment (UE) , personal computers, desktops, mobile phones, cellular phones, smart phones, personal digital assistants (PDAs) , portable computers, tablets, wearable devices, internet of things (IoT) devices, Ultra-reliable and Low Latency Communications (URLLC) devices, Internet of Everything (IoE) devices, machine type communication (MTC) devices, device on vehicle for V2X communication where X means pedestrian, vehicle, or infrastructure / network, devices for Integrated Access and Backhaul (IAB) , Small Data Transmission (SDT) , mobility, Multicast and Broadcast Services (MBS) , positioning, dynamic / flexible duplex in commercial networks, reduced capability (RedCap) , Space borne vehicles or Air borne vehicles in Non-terrestrial networks (NTN) including Satellites and High Altitude Platforms (HAPs) encompassing Unmanned Aircraft Systems (UAS) , eXtended Reality (XR) devices including different types of realities such as Augmented Reality (AR) , Mixed Reality (MR) and Virtual Reality (VR) , the unmanned aerial vehicle (UAV) commonly known as a drone which is an aircraft without any human pilot, devices on high speed train (HST) , or image capture devices such as digital cameras, sensors, gaming devices, music storage and playback appliances, or Internet appliances enabling wireless or wired Internet access and browsing and the like. The ‘terminal device’ can further has ‘multicast / broadcast’ feature, to support public safety and mission critical, V2X applications, transparent IPv4 / IPv6 multicast delivery, IPTV, smart TV, radio services, software delivery over wireless, group communications and IoT applications. It may also incorporate one or multiple Subscriber Identity Module (SIM) as known as Multi-SIM. The term “terminal device” can be used interchangeably with a UE, a mobile station, a subscriber station, a mobile terminal, a user terminal or a wireless device.
[0027] The term “network device” refers to a device which is capable of providing or hosting a cell or coverage where terminal devices can communicate. Examples of a network device include, but not limited to, a Node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a next generation NodeB (gNB) , a transmission reception point (TRP) , a remote radio unit (RRU) , a radio head (RH) , a remote radio head (RRH) , an IAB node, a low power node such as a femto node, a pico node, a reconfigurable intelligent surface (RIS) , Network-controlled Repeaters, and the like.
[0028] The terminal device or the network device may have Artificial intelligence (AI) or Machine learning capability. It generally includes a model which has been trained from numerous collected data for a specific function, and can be used to infer some target information.
[0029] The terminal or the network device may work on several frequency ranges, e.g. FR1 (410 MHz –7125 MHz) , FR2 (24.25GHz to 71GHz) , frequency band larger than 100GHz as well as Tera Hertz (THz) . It can further work on licensed / unlicensed / shared spectrum. The terminal device may have more than one connection with the network devices under Multi-Radio Dual Connectivity (MR-DC) application scenario. The terminal device or the network device can work on full duplex, flexible duplex and cross division duplex modes.
[0030] The network device may have the function of network energy saving, Self-Organizing Networks (SON) / Minimization of Drive Tests (MDT) . The terminal may have the function of power saving.
[0031] The embodiments of the present disclosure may be performed in test equipment, e.g. signal generator, signal analyzer, spectrum analyzer, network analyzer, test terminal device, test network device, channel emulator.
[0032] The embodiments of the present disclosure may be performed according to any generation communication protocols either currently known or to be developed in the future. Examples of the communication protocols include, but not limited to, the first generation (1G) , the second generation (2G) , 2.5G, 2.75G, the third generation (3G) , the fourth generation (4G) , 4.5G, the fifth generation (5G) communication protocols, 5.5G, 5G-Advanced networks, or the sixth generation (6G) networks.
[0033] As used herein, the singular forms ‘a’ , ‘an’ and ‘the’ are intended to include the plural forms as well, unless the context clearly indicates otherwise. The term ‘includes’ and its variants are to be read as open terms that mean ‘includes, but is not limited to. ’ The term ‘based on’ is to be read as ‘at least in part based on. ’ The term ‘some embodiments’ and ‘an embodiment’ are to be read as ‘at least some embodiments. ’ The term ‘another embodiment’ is to be read as ‘at least one other embodiment. ’ The terms ‘first, ’ ‘second, ’ and the like may refer to different or same objects. Other definitions, explicit and implicit, may be included below.
[0034] In some examples, values, procedures, or apparatus are referred to as ‘best, ’ ‘lowest, ’ ‘highest, ’ ‘minimum, ’ ‘maximum, ’ or the like. It will be appreciated that such descriptions are intended to indicate that a selection among many used functional alternatives can be made, and such selections need not be better, smaller, higher, or otherwise preferable to other selections.
[0035] As described above, in label based model performance monitoring methods for UE-based positioning with the AI or ML model at UE side, a ground truth label (i.e., target UE location) is provided for monitoring accuracy of outputs of the AI or ML model. There would be two approaches to obtain the ground truth label. In a first approach, a location server or UE may apply legacy positioning methods to estimate target UE location. In a second approach, a location of a PRU may be used.
[0036] An AI or ML model targets to scenarios with heavy non-line of sight (NLOS) conditions where the legacy positioning methods used in the first approach usually cannot provide an accurate target UE location and the monitoring results can not reflect the AI or ML model performance. Therefore, a PRU with a known location in the second approach should be used to help perform model performance monitoring. However, a PRU may not be always available for a target UE considering the PRU and the target UE should have same or approximate environments or channel conditions. Thus, to perform model performance monitoring in a general situation, a comprehensive procedures should be designed.
[0037] In view of the above, embodiments of the present disclosure provide a solution for communication. In this solution, a location server receives, from a terminal device, at least one first reference output of at least one AI or ML model and determines a second reference output of the at least one AI or ML model based on the at least one first reference output. In turn, the location server determines, based on the second reference output, whether a positioning reference unit (PRU) associated with the terminal device is available for model performance monitoring of the at least one AI or ML model at the terminal device. Then, the location server transmits, to the terminal device, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable. With this solution, model performance monitoring may be performed in a general situation.
[0038] Hereinafter, principle of the present disclosure will be described with reference to Figs. 1A to 11.
[0039] Fig. 1A illustrates a schematic diagram of an example communication network 100 in which embodiments of the present disclosure can be implemented. As shown in Fig. 1A, the communication network 100 comprises a terminal device or a positioning reference unit (PRU) 110, a network device 120 and a location server 130.
[0040] In some embodiments, the location server 130 may be a physical or logical entity that manages positioning for a target device by obtaining measurements and other location information from one or more positioning units and providing assistance data to positioning units to help determine this. The location server 130 may also compute or verify the final location estimate.
[0041] In some embodiments, the location server 130 may comprise one of the following: an Enhanced Serving Mobile Location Centre (E-SMLC) , a Location Management Function (LMF) or Secure User Plane Location (SUPL) Location Platform (SLP) .
[0042] In some embodiments, the terminal device or PRU 110 may communicate with the location server 130 based on LPP. LPP is used point-to-point between the location server 130 and a target device in order to position the target device using position-related measurements obtained by one or more reference sources. For example, the target device may comprise a UE or SUPL Enabled Terminal (SET) .
[0043] In some embodiments, the network device 120 may communicate with the location server 130 based on a New Radio (NR) Positioning Protocol A (NRPPa) . The NRPPa procedure modules are divided into two modules as follows: NRPPa Location Information Transfer Procedures and NRPPa Management Procedures.
[0044] The NRPPa Location Information Transfer Procedures module contains procedures used to handle the transfer of positioning related information between NG-RAN Node and LMF. The Management Procedures module contains procedures that are not related specifically to positioning, i.e., error handling.
[0045] It is to be understood that the number of network devices and terminal devices is only for the purpose of illustration without suggesting any limitations. The communication network 100 may comprise any suitable number of network devices and terminal devices adapted for implementing embodiments of the present disclosure.
[0046] Fig. 1B illustrates another example communication network 100B in which embodiments of the present disclosure can be implemented. Specifically, Fig. 1B illustrates UE Positioning Overall Architecture applicable to NG-RAN.
[0047] As shown in Fig. 1B, the communication network 100B comprises the terminal device 110 and the location server 130 in Fig. 1A. The communication network 100B also comprises a NG-RAN node 140. The NG-RAN node 140 may comprise an ng-eNB 140-1 and a gNB 140-2. Each of the ng-eNB 140-1 and the gNB 140-2 may act as the terminal device 120 in Fig. 2. The location server 130 is implemented as an LMF 130.
[0048] The communication network 100B also comprises an access and mobility management functions (AMF) 150, an Enhanced Serving Mobile Location Centre (E-SMLC) 160 and a SUPL Location Platform (SLP) 170.
[0049] The AMF 150 receives a request for some location service associated with a particular target UE from another entity (e.g., GMLC or UE) or the AMF 150 itself decides to initiate some location service on behalf of a particular target UE (e.g., for an IMS emergency call from the UE) as described in TS 23.502 and TS 23.273. The AMF 150 then sends a location services request to an LMF 130. The LMF 130 processes the location services request which may include transferring assistance data to the target UE to assist with UE-based and / or UE-assisted positioning and / or may include positioning of the target UE. The LMF 130 then returns the result of the location service back to the AMF 150 (e.g., a position estimate for the UE. In the case of a location service requested by an entity other than the AMF 150 (e.g., a GMLC or UE) , the AMF 150 returns the location service result to this entity.
[0050] The NG-RAN node 140 may control several TRPs / TPs, such as remote radio heads, or DL-PRS-only TPs for support of PRS-based TBS.
[0051] The LMF 130 may have a proprietary signalling connection to the E-SMLC 160 which may enable the LMF 130 to access information from E UTRAN (e.g. to support the OTDOA for E-UTRA positioning method using downlink measurements obtained by a target UE of signals from eNBs and / or PRS-only TPs in E-UTRAN) . Details of the signalling interaction between the LMF 130 and E-SMLC 160 are outside the scope of this specification.
[0052] The LMF 130 may have a proprietary signalling connection to the SLP 170. The SLP 170 is the SUPL entity responsible for positioning over the user plane.
[0053] Fig. 1C illustrates a signaling chart illustrating an example process 100C for Location Service Support by NG-RAN in accordance with some embodiments of the present disclosure. For the purpose of discussion, the process 200 will be described with reference to Fig. 1B.
[0054] In the process 100C, when the AMF 150 receives a Location Service Request in case of the UE 110 is in CM-IDLE state, the AMF 150 performs a network triggered service Request as defined in TS 23.502 and TS 23.273 in order to establish a signalling connection with the UE 110 and assign a specific serving gNB or ng-eNB. The UE 110 is assumed to be in connected mode before the beginning of the flow shown in Fig. 1C; that is, any signalling that might be required to bring the UE 110 to connected mode prior to step 1a is not shown. The signalling connection may, however, be later released (e.g. by the NG-RAN node 140 as a result of signalling and data inactivity) while positioning is still ongoing.
[0055] At step 1a, some entity in the 5GC (e.g. GMLC) requests some location service (e.g. positioning) for a target UE 110 to the serving AMF 150.
[0056] Alternatively, at step 1b, the serving AMF 150 for a target UE 110 determines the need for some location service (e.g. to locate the UE 110 for an emergency call) .
[0057] Alternatively, at step 1c, the UE 110 Requests some location service (e.g. positioning or delivery of assistance data) to the serving AMF 150 at the NAS level.
[0058] At step 2, the AMF 150 transfers the location service Request to an LMF 130.
[0059] At step 3a, the LMF 130 instigates location procedures with the serving and possibly neighbouring ng-eNB or gNB in the NG-RAN –e.g. to obtain positioning measurements or assistance data.
[0060] In addition to step 3a or instead of step 3a, the LMF 130 instigates, at step 3b, location procedures with the UE 110 –e.g. to obtain a location estimate or positioning measurements or to transfer location assistance data to the UE 110.
[0061] At step 4, the LMF 130 provides a location service response to the AMF 150 and includes any needed results –e.g. success or failure indication and, if Requested and obtained, a location estimate for the UE 110.
[0062] If step 1a was performed, the AMF 150 returns, at step 5a, a location service response to the 5GC entity in step 1a and includes any needed results –e.g. a location estimate for the UE 110.
[0063] If step 1b occurred, the AMF 150 uses, at step 5b, the location service response received in step 4 to assist the service that triggered this in step 1b (e.g. may provide a location estimate associated with an emergency call to a GMLC) .
[0064] If step 1c was performed, the AMF 150 returns, at step 5c, a location service response to the UE 110 and includes any needed results –e.g. a location estimate for the UE 110.
[0065] Location procedures applicable to NG-RAN occur in steps 3a and 3b in Fig. 1C and are defined in greater detail in this specification. Other steps in Fig. 1C are applicable only to the 5GC and are described in greater detail and in TS 23.502 and TS 23.273.
[0066] Steps 3a and 3b can involve the use of different position methods to obtain location related measurements for a target UE 110 and from these compute a location estimate and possibly additional information like velocity. Positioning methods supported in this release are summarized in clause 4.3 and described in detail in clause 8 of TS 38.305.
[0067] Fig. 1D illustrates a signaling chart illustrating an example process 100D for SLPP PDU transfer between UEs in accordance with some embodiments of the present disclosure.
[0068] As shown in Fig. 1D, at step 1, the TX UE 110-1 determines to send an SLPP message to the RX UE 110-2 as part of some SLPP positioning activity (e.g., to instigate a SLPP procedure or in response to a previously received SLPP message) . The SLPP layer in the TX UE 110-1 determines the SLPP PDU and the application layer ID for the peer UE.
[0069] At step 2, the TX UE 110-1 establishes a unicast link with the peer UE using the application layer ID received from step 1, if not already established, as specified in TS 23.287 for V2X capable UEs and TS 23.304 for ProSe capable UEs.
[0070] At step 3, the TX UE 110-1 determines the transport configurations for the SLPP PDU. For V2X capable UEs, policies and parameters defined in TS 23.287 clause 5.1.2.1 are used to determine the corresponding transport configurations for the SLPP PDU. For ProSe capable UEs, policies and parameters defined in TS 23.304 clause 5.1.3.1 are used to determine the corresponding transport configurations for the SLPP PDU.
[0071] At step 4, the SLPP PDU is transmitted via the established unicast link according to the procedures for V2X communication over PC5 reference point defined in TS 23.287 or according to the procedures for 5G ProSe direct communication defined in TS 23.304.
[0072] Fig. 2 illustrates a signaling chart illustrating an example process 200 for communications in accordance with some embodiments of the present disclosure. For the purpose of discussion, the process 200 will be described with reference to Fig. 1A or 1B. The process 200 may involve the terminal device 110 and the location server 130 in Fig. 1A or 1B.
[0073] As shown in Fig. 2, the terminal device 110 transmits 210 at least one first reference output of at least one AI or ML model to the location server 130.
[0074] In some embodiments, each of the at least one first reference output indicates a predicted location of the terminal device 110.
[0075] In some embodiments, the at least one first reference output of at least one AI or ML model may be one or more reference output of a single AI or ML model in the terminal device 110.
[0076] Alternatively, in some embodiments, the at least one first reference output of at least one AI or ML model may be multiple reference outputs of multiple AI or ML models. Each of the multiple AI or ML models has one or more reference output.
[0077] In some embodiments, the at least one first reference output of at least one AI or ML model may comprise the latest output or periodically determined output from model inference of the terminal device 110.
[0078] In some embodiments, the terminal device 110 may initiate a model performance monitoring procedure. The model performance monitoring procedure may be defined as a procedure that monitors performance of an AI or ML model. For example, the model performance monitoring procedure may be initiated by the terminal device 110 based on a request for model performance monitoring received 240 from the location server 130 or initiated by the terminal device 110 itself.
[0079] In turn, the terminal device 110 may transmit a request for information about a PRU associated with the terminal device 110 to the location server 130. The request may comprise the at least one first reference output.
[0080] Alternatively, in some embodiments, the terminal device 110 may transmit the at least one first reference output together or within or separate from the request for information about the PRU together.
[0081] In some embodiments, the information about the PRU may comprise downlink channel measurement of the PRU (e.g., PRU performs DL-PRS measurements) and the corresponding location of the PRU which will be used by the terminal device 110 for model performance monitoring.
[0082] In some embodiments, the information about the PRU may further comprise line of sight (LOS) or NLOS indication of the PRU.
[0083] Upon receiving the request for information about a PRU associated with the terminal device 110, the location server 130 may together or additionally receive at least one first reference output of at least one AI or ML model. The location server 130 further determines 220 a second reference output of the at least one AI or ML model based on the at least one first reference output.
[0084] In turn, the location server 130 determines 230, based on the second reference output, whether the PRU associated with the terminal device 110 is available for model performance monitoring of the at least one AI or ML model at the terminal device 110.
[0085] In some embodiments, the location server 130 may determine the second reference output as one of the at least one first reference output. In such embodiments, the location server 130 may determine the second reference output as a reference output which is last received or known from the terminal device 110. For example, the location server 130 may determine the second reference output as a reference output within the request for the information about the PRU or a reference output received together with the request for the information about the PRU. For another example, the location server 130 may determine the second reference output as a reference output from a previously received LPP message from the terminal device 110.
[0086] In some embodiments, the location server 130 may transmit 250, to the terminal device 110, a request for an updated reference output of the at least one AI or ML model. Then, the location server 130 may receive 260 the updated reference output from the terminal device 110. In turn, the location server 130 may determine the second reference output as the updated reference output.
[0087] In some embodiments, the updated reference output may be based on a last, next or new model inference at the terminal device 110.
[0088] Alternatively or additionally, if a difference (T2-T1) between current time (T2) and first time (T1) associated with a last received one of the at least one first reference output exceeds a time threshold, the location server 130 may transmit the request for the updated reference output to the terminal device 110.
[0089] Alternatively, in some embodiments, the location server 130 may request the terminal device 110 to perform a legacy positioning methods to obtain an estimated location of the terminal device 110 and report the estimated location to the location server 130 or the location server 130 performs legacy positioning methods to obtain the estimated location of the terminal device 110. In such embodiments, if the terminal device 110 determines that the estimated location and the reference output are close enough, the terminal device 110 may determine that the model performance monitoring procedure has a good model performance monitoring output and stop the current model performance monitoring procedure based on the information about the PRU. Alternatively, the location server 130 may determine the second reference output to determine whether the PRU is available based on the second reference output and / or the estimated location of the terminal device 110.
[0090] In some embodiments, if a distance between a first position of the PRU and a second position indicated by the second reference output is below a distance threshold, the location server 130 may determine the PRU associated with the terminal device 110 is available. In other words, if the first position of the PRU is close enough to the second position indicated by the second reference output, the location server 130 may determine the PRU associated with the terminal device 110 is available.
[0091] In some embodiments, the distance may be one of the following: a two-dimension distance, three-dimension distance or radio distance. For example, the location server 130 may request the terminal device 110 or the PRU to measure their respective Reference Signal Receiving Power (RSRP) from each other and report to the location server 130 as the radio distance.
[0092] Additionally or alternatively, if there are multiple associated PRU available, the location server 130 may determines a PRU with NLOS condition as the PRU associated with the terminal device 110.
[0093] Alternatively, if the estimated location of the terminal device 110 and the second reference output are close enough, the location server 130 may determine the PRU with LOS indicator as the PRU associated with the terminal device 110. If the estimated location of the terminal device 110 and the second reference output are not close enough, the location server 130 may determine the PRU with NLOS indicator as the PRU associated with the terminal device 110.
[0094] Alternatively, the terminal device 110 may report NLOS indicator or LOS indicator from model inference to the location server 130, and the location server 130 may determine a PRU with same NLOS / LOS condition as the terminal device 110 to be the PRU associated with the terminal device 110.
[0095] Fig. 3 illustrates an example of determining a PRU associated with the terminal device 110 in accordance with some embodiments of the present disclosure.
[0096] As shown in Fig. 3, the terminal device 110 transmits a first reference output of an AI or ML model to the location server 130. The first reference output is represented by R and indicates a location 310 of the terminal device 110. The terminal device 110 is actually located at a location 312. The location server 130 determines a second reference output of the AI or ML model based on the first reference output of the AI or ML model. The location server 130 determines an area or region 314 with a radius D which is based on the second reference output. The location server 130 determines a PRU 320 in the area or region 314 as a PRU associated with the terminal device 110.
[0097] In turn, the location server 130 may cause information about the PRU 320 to be transmitted from the PRU 320 to the terminal device 110 via a unicast sidelink (SL) directly. This will be described with reference to Fig. 4.
[0098] Alternatively, the location server 130 may forward the information about the PRU 320, which is received from the PRU 320, to the terminal device 110. This will be described with reference to Fig. 5.
[0099] Fig. 4 illustrates a signaling chart illustrating an example process 400 for communications in accordance with some embodiments of the present disclosure. The process 400 may be considered as an example implementation of the process 200. For the purpose of discussion, the process 400 will be described with reference to Figs. 1 and 3. The process 400 may involve the terminal device 110, the network device 120 and the location server 130 in Fig. 1A and the PRU 320 in Fig. 3.
[0100] As shown in Fig. 4, the location server 130 determines 410 the PRU 320 associated with the terminal device 110 is available.
[0101] The location server 130 configures and triggers 420 reference signal transmission and channel measurement to obtain information about the PRU 320.
[0102] The network device 120 performs 430 the reference signal transmission to the PRU 320.
[0103] The PRU 320 performs 440 the channel measurement.
[0104] In some embodiments, the terminal device 110 and the PRU 320 are in approximate locations. If the PRU 320 and the terminal device 110 are V2X capable or ProSe capable UEs, the terminal device 110 may transmit 450 V2X or ProSe capability information about the terminal device 110 to the location server 130, and the PRU 320 may transmit 460 V2X or ProSe capability information about the PRU 320 to the location server 130.
[0105] In turn, the location server 130 (or associated AMF, core network, high layer or application layer) may trigger following Sidelink Positioning Protocol (SLPP) protocol data unit (PDU) procedure from the PRU 320 to the terminal device 110 directly.
[0106] Specifically, the SLPP PDU comprises the information about the PRU 320 with a traffic priority assigned by the location server 130. The PRU 320 establishes a unicast sidelink with the terminal device 110 if it is not already established. The PRU 320 transmits 470 the SLPP PDU to the terminal device 110 via the established unicast sidelink.
[0107] Upon receiving the information about the PRU 320, the terminal device 110 performs 480 the model performance monitoring based on the information about the PRU 320.
[0108] Fig. 5 illustrates a signaling chart illustrating an example process 500 for communications in accordance with some embodiments of the present disclosure. The process 500 may be considered as an example implementation of the process 200. For the purpose of discussion, the process 500 will be described with reference to Figs. 1 and 3. The process 500 may involve the terminal device 110, the network device 120 and the location server 130 in Fig. 1A and the PRU 320 in Fig. 3.
[0109] Actions 410, 420, 430, 440 and 480 in the process 500 are the same as those in the process 400. Details of these actions are omitted for brevity.
[0110] The process 500 is different from the process 400 in actions 510, 520 and 530.
[0111] Specifically, the PRU 320 transmits 510 the channel measurements of the PRU 320 and a location of the PRU 320 to the location server 130.
[0112] The location server 130 determines 520 the information about the PRU 320 based on the channel measurements of the PRU 320 and the location of the PRU 320.
[0113] The location server 130 transmits 530 the information about the PRU 320 to the terminal device 110.
[0114] In some embodiments, if the location server 130 determines that the PRU associated with the terminal device 110 is unavailable, the location server 130 may transmit an indication to the terminal device 110. The indication indicates at least one of the following: the PRU associated with the terminal device 110 is unavailable, information about the PRU is unavailable, or PRU based model performance monitoring is not applicable. For example, the location server 130 may transmit an LPP message comprising the indication. The LPP message may be transmitted per model, e.g., with model ID or associated ID.
[0115] The terminal device 110 may perform one of the following based on the indication: model switch, model deactivation, or model fallback. For example, the terminal device 110 may perform one of the following upon receiving the indication: model switch, model deactivation, or model fallback.
[0116] In some embodiments, the terminal device 110 may perform model fallback only if all models in the terminal device 110 are informed with no associated PRU information.
[0117] Alternatively, the terminal device 110 may start a timer upon receiving the indication. The terminal device 110 may attempt to perform a subsequent model performance monitoring procedure during the timer is running. During the timer is running, the terminal device 110 instigates a subsequent model performance monitoring procedure by itself or based on a request from the location server 130. The terminal device 110 may stop the timer once a model performance monitoring is done and / or a good monitoring output is obtained. Otherwise, the terminal device 110 may perform model switch, model deactivation, or model fallback when the timer expires.
[0118] Alternatively, the terminal device 110 may maintain a counter to accumulate the received LPP message and / or bad model performance monitoring output.
[0119] The terminal device 110 may increase the counter by one if the terminal device 110 received the above LPP message or a bad model performance monitoring output is obtained.
[0120] Once a good model performance monitoring output is obtained, the terminal device 110 may reset the counter.
[0121] The terminal device 110 may determine to perform model switch, model deactivation, or model fallback if the counter reaches a threshold.
[0122] In some embodiments, there may be multiple ongoing model performance monitoring procedures.
[0123] Alternatively, in some embodiments, the terminal device 110 may perform, based on the indication, the model performance monitoring based on a label (i.e., location of the terminal device 110) derived from a positioning method. For example, the terminal device 110 may perform, based on the indication, the model performance monitoring based on a label (i.e., location of the terminal device 110) derived from a legacy positioning method.
[0124] In some embodiments, under the condition that the location of the terminal device 110 derived from legacy positioning methods has high accuracy, the terminal device 110 may perform the model performance monitoring based on the label derived from legacy positioning methods. For example, if LOS possibility is relative high, or measurements for positioning has relative high quality, from relative high accurate NR-independent positioning methods, e.g., Global Navigation Satellite System (GNSS) , the terminal device 110 may perform the model performance monitoring based on the label derived from legacy positioning methods
[0125] Alternatively, in some embodiments, the terminal device 110 may perform, based on the indication, label-free model performance monitoring. For example, the terminal device 110 may perform input / output distribution, relative displacement, inference output inconsistency based model performance monitoring.
[0126] Fig. 6 illustrates a signaling chart illustrating an example process 600 for communications in accordance with some embodiments of the present disclosure. The process 600 may be considered as an example implementation of the process 200. For the purpose of discussion, the process 600 will be described with reference to Figs. 1 and 3. The process 600 may involve the terminal device 110 and the location server 130 in Fig. 1A.
[0127] Actions 210, 220 and 230 in the process 600 are the same as those in the process 200. Details of these actions are omitted for brevity.
[0128] The process 600 is different from the process 200 in actions 610 and 620.
[0129] Specifically, the location server 130 transmits 610, to the terminal device 110, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.
[0130] In some embodiments, if the location server 130 determines that the PRU associated with the terminal device 110 is unavailable, the location server 130 may transmit the at least one reference location to the terminal device 110.
[0131] If one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, the terminal device 110 transmits 620, to the location server 130, a request for the model performance monitoring.
[0132] Fig. 7 illustrates an example of reference locations in accordance with some embodiments of the present disclosure.
[0133] As shown in Fig. 7, the terminal device 110 transmits a first reference output of an AI or ML model to the location server 130. The first reference output is represented by R and indicates a location 710 of the terminal device 110. The terminal device 110 is actually located at a location 712. The location server 130 determines the second reference output of the AI or ML model as the first reference output of the AI or ML model. The location server 130 determines that the PRU associated with the terminal device 110 is unavailable based on the second reference output.
[0134] In turn, the location server 130 transmits to the terminal device 110, at least one reference location where a reference PRU 720 is available or information about the reference PRU 720 is available or the model performance monitoring is applicable.
[0135] For example, the at least one reference location may comprise at least one of the following: a location 722, a location 724, or a location 726.
[0136] At the location 722, information about the reference PRU 720 is available or the model performance monitoring is applicable. For example, the location 722 is the closest to the location 710.
[0137] The location 724 is a location of the reference PRU 720. For example, the location 724 is the closest to the location 710.
[0138] The location 726 is a region where the information about the reference PRU 720 is available. The region 726 is the closest to the location 710.
[0139] If one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, the terminal device 110 transmits, to the location server 130, a request for the model performance monitoring.
[0140] Alternatively, in some embodiments, if the PRU associated with the terminal device 110 is unavailable, the terminal device 110 may transmit, to the location server 130, a request for further model performance monitoring after a configured time duration. For example, the terminal device 110 may transmit, to the location server 130, the request for further model performance monitoring periodically.
[0141] Alternatively, in some embodiments, the location server 130 may trigger further model performance monitoring by transmitting, to the terminal device 110, a request for an updated reference output of the at least one AI or ML model. For example, because of a new deployed PRU or moving PRU, the location server 130 may trigger the further model performance monitoring. The terminal device 110 may transmit an updated reference output of an AI or ML model to the location server 130 upon receiving the request. The location server 130 may determine a PRU associated with the terminal device 110 based on the updated reference output of the AI or ML model.
[0142] Fig. 8 illustrates a signaling chart illustrating an example process 800 for communications in accordance with some embodiments of the present disclosure. For the purpose of discussion, the process 800 will be described with reference to Fig. 1A. The process 800 may involve the terminal device 110 and the location server 130 in Fig. 1A.
[0143] As shown in Fig. 8, the location server 130 transmits 810, to the terminal device 110, at least one reference location where a PRU is available or information about the reference PRU is available or model performance monitoring of at least one AI or ML model is applicable. Some embodiments of the at least one reference location has been described with reference to Fig. 7. Detail of such embodiments are omitted for brevity.
[0144] If one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, the terminal device 110 transmits 820, to the location server 130, a request for the model performance monitoring.
[0145] The terminal device 110 transmits 830 the at least one first reference output to the location server 130.
[0146] Upon receiving the at least one first reference output of at least one AI or ML model, the location server 130 determines 840 a second reference output of the at least one AI or ML model based on the at least one first reference output.
[0147] In turn, the location server 130 determines 850, based on the second reference output, whether the PRU associated with the terminal device 110 is available for model performance monitoring of the at least one AI or ML model at the terminal device 110.
[0148] The actions 840 and 850 are the same as the actions 220 and 230 in the process 200. Details of these actions are omitted for brevity.
[0149] If the PRU associated with the terminal device 110 is available, the location server 130 may transmit 860 information about the PRU associated with the terminal device 110 to the terminal device 110. Alternatively, the location server 130 may cause the information about the PRU to be transmitted from the PRU to the terminal device 110.
[0150] If information about a PRU associated with the terminal device 110 is received from the location server 130 or the PRU, the terminal device 110 performs 870 the model performance monitoring based on the information about the PRU.
[0151] Fig. 9 illustrates a flowchart of an example method in accordance with some embodiments of the present disclosure. In some embodiments, the method 900 can be implemented at a location server, such as the location server 130 as shown in Fig. 1A. For the purpose of discussion, the method 900 will be described with reference to Fig. 1A.
[0152] At block 910, the location server 130 receives, from a terminal device 110, at least one first reference output of at least one AI or ML model.
[0153] At block 920, the location server 130 determines a second reference output of the at least one AI or ML model based on the at least one first reference output.
[0154] At block 930, the location server 130 determines, based on the second reference output, whether a PRU associated with the terminal device 110 is available for model performance monitoring of the at least one AI or ML model at the terminal device 110.
[0155] At block 940, the location server 130 transmits, to the terminal device 110, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.
[0156] In some embodiments, receiving the at least one first reference output may comprise: receiving a request for information about the PRU from the terminal device 110, wherein the request comprises the at least one first reference output; or receiving the request together with the at least one first reference output from the terminal device 110.
[0157] In some embodiments, determining the second reference output may comprise: determining the second reference output as one of the at least one first reference output.
[0158] In some embodiments, determining the second reference output may comprise: transmitting, to the terminal device 110, a request for an updated reference output of the at least one AI or ML model; receiving the updated reference output from the terminal device 110; and determining the second reference output as the updated reference output.
[0159] In some embodiments, transmitting the request for the updated reference output determining the second reference output may comprise: based on determining that a difference between current time and first time associated with a last received one of the at least one first reference output exceeds a time threshold, transmitting the request for the updated reference output.
[0160] In some embodiments, determining the PRU associated with the terminal device 110 is available may comprise: based on determining that a distance between a first position of the PRU and a second position indicated by the second reference output is below a distance threshold, determining the PRU associated with the terminal device 110 is available.
[0161] In some embodiments, the method 900 further comprises: based on determining that the PRU associated with the terminal device 110 is available, cause information about the PRU to be transmitted from the PRU to the terminal device 110.
[0162] In some embodiments, the method 900 further comprises: based on determining that the PRU associated with the terminal device 110 is available, causing the PRU to establish a unicast sidelink with the terminal device 110; and causing information about the PRU to be transmitted from the PRU to the terminal device 110 via the unicast sidelink.
[0163] In some embodiments, the information about the PRU is associated with a traffic priority.
[0164] In some embodiments, the method 900 further comprises: based on determining that the PRU associated with the terminal device 110 is unavailable, transmitting an indication to the terminal device 110, wherein the indication indicates at least one of the following: the PRU associated with the terminal device 110 is unavailable, information about the PRU is unavailable, or PRU based model performance monitoring is not applicable.
[0165] In some embodiments, the method 900 further comprises: based on determining that the PRU associated with the terminal device 110 is unavailable, triggering further model performance monitoring by transmitting, to the terminal device 110, a request for an updated reference output of the at least one AI or ML model.
[0166] Fig. 10 illustrates a flowchart of an example method in accordance with some embodiments of the present disclosure. In some embodiments, the method 1000 can be implemented at a terminal device, such as the terminal device 110 as shown in Fig. 1A. For the purpose of discussion, the method 1000 will be described with reference to Fig. 1A.
[0167] At block 1010, the terminal device 110 receives, from the location server 130, at least one reference location where a reference PRU is available or information about the reference PRU is available or model performance monitoring of at least one AI or ML model is applicable.
[0168] At block 1020, based on determining that one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, the terminal device 110 transmits, to the location server 130, a request for the model performance monitoring.
[0169] At block 1030, the terminal device 110 transmits the at least one first reference output to the location server 130.
[0170] At block 1040, based on determining that information about a PRU associated with the terminal device 110 is received from the location server 130 or the PRU, the terminal device 110 performs the model performance monitoring based on the information about the PRU.
[0171] In some embodiments, transmitting the at least one first reference output may comprise: transmitting a request for information about the PRU to the location server 130, wherein the request comprises the at least one first reference output; or transmitting the request together with the at least one first reference output to the location server 130.
[0172] In some embodiments, the method 100 may further comprise: receiving, from the location server 130, a request for an updated reference output of the at least one AI or ML model; and transmitting the updated reference output to the location server 130 based on the request.
[0173] In some embodiments, the information about the PRU is associated with a traffic priority.
[0174] In some embodiments, the method 100 may further comprise: receive an indication from the location server 130, wherein the indication indicates at least one of the following: the PRU associated with the terminal device 110 is unavailable, information about the PRU is unavailable, or PRU based model performance monitoring is not applicable.
[0175] In some embodiments, the method 100 may further comprise: performing one of the following based on the indication: model switch, model deactivation, or model fallback.
[0176] In some embodiments, performing the model switch, model deactivation, or model fallback may comprise performing the model switch, model deactivation, or model fallback upon receiving the indication.
[0177] In some embodiments, performing the model switch, model deactivation, or model fallback may comprise: starting a timer upon receiving the indication; and performing the model switch, model deactivation, or model fallback based on determining that the timer expires.
[0178] In some embodiments, the method 100 may further comprise: performing, based on the indication, the model performance monitoring based on a label derived from a positioning method; or performing, based on the indication, label-free model performance monitoring.
[0179] In some embodiments, the method 100 may further comprise: transmitting, to the location server 130, a request for further model performance monitoring after a configured time duration.
[0180] Fig. 11 is a simplified block diagram of a device 1100 that is suitable for implementing embodiments of the present disclosure. The device 1100 can be considered as a further example embodiment of the terminal device 110 or the location server 130 as shown in Fig. 1. Accordingly, the device 1100 can be implemented at or as at least a part of the terminal device 110 or the location server 130.
[0181] As shown, the device 1100 includes a processor 1110, a memory 1120 coupled to the processor 1110, a suitable transceiver 1140 coupled to the processor 1110, and a communication interface coupled to the transceiver 1140. The memory 1110 stores at least a part of a program 1130. The transceiver 1140 may be for bidirectional communications or a unidirectional communication based on requirements. The transceiver 1140 may include at least one of a transmitter 1142 and a receiver 1144. The transmitter 1142 and the receiver 1144 may be functional modules or physical entities. The transceiver 1140 has at least one antenna to facilitate communication, though in practice an Access Node mentioned in this application may have several ones. The communication interface may represent any interface that is necessary for communication with other network elements, such as X2 / Xn interface for bidirectional communications between eNBs / gNBs, S1 / NG interface for communication between a Mobility Management Entity (MME) / Access and Mobility Management Function (AMF) / SGW / UPF and the eNB / gNB, Un interface for communication between the eNB / gNB and a relay node (RN) , or Uu interface for communication between the eNB / gNB and a terminal device.
[0182] The components included in the apparatuses and / or devices of the present disclosure may be implemented in various manners, including software, hardware, firmware, or any combination thereof. In one embodiment, one or more units may be implemented using software and / or firmware, for example, machine-executable instructions stored on the storage medium. In addition to or instead of machine-executable instructions, parts or all of the units in the apparatuses and / or devices may be implemented, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include Field-programmable Gate Arrays (FPGAs) , Application-specific Integrated Circuits (ASICs) , Application-specific Standard Products (ASSPs) , System-on-a-chip systems (SOCs) , Complex Programmable Logic Devices (CPLDs) , and the like.
Claims
1.A location server, comprising:a processor configured to cause the location server to:receive, from a terminal device, at least one first reference output of at least one artificial intelligence (AI) or machine learning (ML) model;determine a second reference output of the at least one AI or ML model based on the at least one first reference output;determine, based on the second reference output, whether a positioning reference unit (PRU) associated with the terminal device is available for model performance monitoring of the at least one AI or ML model at the terminal device; andtransmit, to the terminal device, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.2.The location server of claim 1, wherein the location server is caused to receive the at least one first reference output by:receiving a request for information about the PRU from the terminal device, wherein the request comprises the at least one first reference output; orreceiving the request together with the at least one first reference output from the terminal device.3.The location server of claim 1, wherein the location server is caused to determine the second reference output by:determining the second reference output as one of the at least one first reference output.4.The location server of claim 1, wherein the location server is caused to determine the second reference output by:transmitting, to the terminal device, a request for an updated reference output of the at least one AI or ML model;receiving the updated reference output from the terminal device; anddetermining the second reference output as the updated reference output.5.The location server of claim 4, wherein the location server is caused to transmit the request for the updated reference output by:based on determining that a difference between current time and first time associated with a last received one of the at least one first reference output exceeds a time threshold, transmitting the request for the updated reference output.6.The location server of claim 1, wherein the location server is caused to determine the PRU associated with the terminal device is available by:based on determining that a distance between a first position of the PRU and a second position indicated by the second reference output is below a distance threshold, determining the PRU associated with the terminal device is available.7.The location server of claim 1, wherein the location server is further caused to:based on determining that the PRU associated with the terminal device is available, cause information about the PRU to be transmitted from the PRU to the terminal device.8.The location server of claim 1, wherein the location server is further caused to:based on determining that the PRU associated with the terminal device is available, cause the PRU to establish a unicast sidelink with the terminal device; andcause information about the PRU to be transmitted from the PRU to the terminal device via the unicast sidelink.9.The location server of claim 7 or 8, wherein the information about the PRU is associated with a traffic priority.10.The location server of claim 1, wherein the location server is further caused to:based on determining that the PRU associated with the terminal device is unavailable, transmit an indication to the terminal device, wherein the indication indicates at least one of the following:the PRU associated with the terminal device is unavailable,information about the PRU is unavailable, orPRU based model performance monitoring is not applicable.11.The location server of claim 1 or 10, wherein the location server is further caused to:based on determining that the PRU associated with the terminal device is unavailable, trigger further model performance monitoring by transmitting, to the terminal device, a request for an updated reference output of the at least one AI or ML model.12.A terminal device, comprising:a processor configured to cause the terminal device to:receive, from a location server, at least one reference location where a reference positioning reference unit (PRU) is available or information about the reference PRU is available or model performance monitoring of at least one artificial intelligence (AI) or machine learning (ML) model is applicable;based on determining that one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, transmit, to the location server, a request for the model performance monitoring;transmit the at least one first reference output to the location server; andbased on determining that information about a PRU associated with the terminal device is received from the location server or the PRU, perform the model performance monitoring based on the information about the PRU.13.The terminal device of claim 12, wherein the terminal device is caused to transmit the at least one first reference output by:transmitting a request for information about the PRU to the location server, wherein the request comprises the at least one first reference output; ortransmitting the request together with the at least one first reference output to the location server.14.The terminal device of claim 12, wherein the terminal device is further caused to:receiving, from the location server, a request for an updated reference output of the at least one AI or ML model; andtransmitting the updated reference output to the location server based on the request.15.The terminal device of claim 12, wherein the information about the PRU is associated with a traffic priority.16.The terminal device of claim 12, wherein the terminal device is further caused to:receive an indication from the location server, wherein the indication indicates at least one of the following:the PRU associated with the terminal device is unavailable,information about the PRU is unavailable, orPRU based model performance monitoring is not applicable.17.The terminal device of claim 16, wherein the terminal device is further caused to:perform one of the following based on the indication:model switch,model deactivation, ormodel fallback.18.The terminal device of claim 17, wherein the terminal device is caused to perform the model switch, model deactivation, or model fallback upon receiving the indication.19.The terminal device of claim 17, wherein the terminal device is caused to perform the model switch, model deactivation, or model fallback by:starting a timer upon receiving the indication; andperforming the model switch, model deactivation, or model fallback based on determining that the timer expires.20.The terminal device of claim 16, wherein the terminal device is further caused to:perform, based on the indication, the model performance monitoring based on a label derived from a positioning method; orperform, based on the indication, label-free model performance monitoring.21.The terminal device of claim 12 or 16, wherein the terminal device is further caused to:transmit, to the location server, a request for further model performance monitoring after a configured time duration.22.A method for communications, comprising:receiving, from a terminal device, at least one first reference output of at least one artificial intelligence (AI) or machine learning (ML) model;determining a second reference output of the at least one AI or ML model based on the at least one first reference output;determining, based on the second reference output, whether a positioning reference unit (PRU) associated with the terminal device is available for model performance monitoring of the at least one AI or ML model at the terminal device; andtransmitting, to the terminal device, at least one reference location where a reference PRU is available or information about the reference PRU is available or the model performance monitoring is applicable.23.A method for communications, comprising:receiving, from a location server, at least one reference location where a reference positioning reference unit (PRU) is available or information about the reference PRU is available or model performance monitoring of at least one artificial intelligence (AI) or machine learning (ML) model is applicable;based on determining that one of at least one first reference output of the at least one AI or ML model fulfils or overlaps with one of the at least one reference location, transmitting, to the location server, a request for the model performance monitoring;transmitting the at least one first reference output to the location server; andbased on determining that information about a PRU associated with the terminal device is received from the location server or the PRU, performing the model performance monitoring based on the information about the PRU.24.A computer readable medium having instructions stored thereon, the instructions, when executed on at least one processor of a device, causing the device to carry out the method according to claim 22 or 23.
Citation Information
Patent Citations
AI or ML model monitoring method and device, communication equipment and storage medium
CN118303012A
Performance supervision method and apparatus for ai network model, and communication device
WO2024027576A1
Ai / ML model monitoring method and device
WO2024148508A1